palaeogeography, palaeoclimatology, palaeoecology · volcanism and regional to global environmental...

Palaeogeography, Palaeoclimatology, Palaeoecology 321–322 (2012) 88–101

Contents lists available at SciVerse ScienceDirect

Palaeogeography, Palaeoclimatology, Palaeoecology

j ourna l homepage: www.e lsev ie r .com/ locate /pa laeo

Bentonite geochronology, marine geochemistry, and the Great OrdovicianBiodiversification Event (GOBE)

Cara K. Thompson a,⁎, Linda C. Kah a, Ricardo Astini b, Samuel A. Bowring c, Robert Buchwaldt c

a Department of Earth and Planetary Sciences University of Tennessee, Knoxville, TN 37996, United Statesb Laboratorio de Análisis de Cuencas, CICTERRA-CONICET, Universidad Nacional de Córdoba, Córdoba, Argentinac Department of Earth, Atmospheric, and Planetary Sciences Massachusetts Institute of Technology, Cambridge, MA 02138, United States

⁎ Corresponding author at: Department of GeoscieStony Brook, NY 11794, United States.

E-mail address: [email protected] (C.K. T

0031-0182/$ – see front matter © 2012 Elsevier B.V. Alldoi:10.1016/j.palaeo.2012.01.022

a b s t r a c t
a r t i c l e i n f o
Article history:Received 21 March 2011Received in revised form 29 November 2011Accepted 20 January 2012Available online 28 January 2012

Keywords:Middle OrdovicianArgentinaK-bentoniteGeochronologyEnvironmental change

Attribution of Ordovician climate forcing to explosive volcanismand the potential global importance of volcanism inOrdovician biodiversification suggest the necessity of evaluating the relationships between K-bentonite depositionand increasingly high-resolution records of marine biogeochemical change. Globally, Ordovician strata preserve anextensive record of explosive volcanism— including thewidely recognized Lower toMiddleOrdovician FamatinaK-bentonite suite in Argentina and the Upper Ordovician Millbrig–Deicke–Kinnekulle suite of North America andEurope. Here, we present high-resolution ID-TIMS U–Pb zircon ages of K-bentonites from measured sections ofthe San Juan Formation (Talacasto and Cerro La Chilca section) of the Argentine Precordillera. K-bentonites fromthe Argentine Precordillera provide stratigraphically consistent (i.e., younging upward) ages that range from473.45±0.70 Ma to 469.53±0.62 Ma, and constrain the age of a low-magnitude (2‰), globally recorded, negativecarbon-isotope excursion. Evaluation of the timing of K-bentonite deposition in the Argentina Precordillera relativeto marine biostratigraphic and biogeochemical records provides insight into relationships between explosivevolcanism and regional to global environmental change. From a regional standpoint, these ages provide criticaldirect evidence for a Dapingian to earliest Darriwilian age of the upper San Juan Formation at sampled localities.These ages are consistent with carbon-isotope data suggesting that the San Juan Formation in the region of itstype section is coeval with only the base of the often-correlated Table Head Group of western Newfoundland.This data thus highlights the difficulties in using regional biostratigraphic data – particularly within erosionallytruncated or otherwise diachronous units – to define the timeframe of carbon-isotope chemostratigraphy. Newgeochronological data also indicate that a discrete negative carbon-isotope excursion within the San Juan andTable Head formations is correlative to a globally recognized pre-MDICE negative excursion, and indicates thatthis aspect of the marine carbon isotope record can be used as a discrete chronologic marker. San Juan Formationbentonites, however, cannot be discretely correlated with observed, environmentally significant changes in theMiddle Ordovician marine geochemical records of carbon, sulfur, strontium, or sea surface temperature. Theseresults suggest that the extent of volcanism represented by the Famatina bentonite suite was insufficient to affectglobal surface environments and that the relationship between explosive volcanism and environmental changemay not be straightforward as previously suggested.

© 2012 Elsevier B.V. All rights reserved.

1. Introduction

The Ordovician is characterized by a series of profound changes inEarth surface environments. Perhaps most dramatic among thesechanges was the expansion of marine life during the Great OrdovicianBiodiversification Event (GOBE) (Harper, 2006). Over a 25 My periodbeginning in the Early Ordovician, a cascade of diversificationresulted in increased biodiversity at species, genus, and family levels(Droser and Sheehan, 1997; Webby et al., 2004; Harper, 2006). The

nces, Stony Brook University,

hompson).

rights reserved.

peak of the GOBE correlates with a maximum continental dispersion(Scotese and McKerrow, 1990; Owen and Crame, 2002; Webby etal., 2004), elevated pCO2 (Berner and Kothavala, 2001; Herrmann etal., 2004), and global greenhouse climates. These conditions resultedin the largest area of tropical marine shelves in the Phanerozoic(Walker et al., 2002; Servais et al., 2009) and the potential forsubstantial diversification as a result of ecological interaction withinan expanded ecosystem space (Droser and Finnegan, 2003; Harper,2006; Achab and Paris, 2007; Servais et al., 2009; Servais et al.,2010). Yet there currently remains no single explanation for theGOBE (see Servais et al., 2009, 2010 for review), and a variety ofpotential causes – including global cooling (Trotter et al., 2008;Zhang et al., 2010), changes in nutrient flux (Cardenas and Harries,2010) and resulting expansion of phytoplankton (Servais et al.,

http://dx.doi.org/10.1016/j.palaeo.2012.01.022

mailto:[email protected]

http://dx.doi.org/10.1016/j.palaeo.2012.01.022

http://www.sciencedirect.com/science/journal/00310182

30°

60°

EQ

Gondwana

Baltica

Laurentia

Precordillera

Siberia

Antarctica

Australia

India

Avalonianarc

IapetusOcean

SouthAmerica

Famatinamagmatic arc

Fig. 1. Paleogeographic reconstruction of the Middle Ordovician (c. 470 Ma; modifiedfrom Astini et al., 2007). The Famatina magmatic arc has been identified (Fanning etal., 2004) as the source of voluminous ash fall deposits in the upper San Juan andlower Gualcamayo formations.

89C.K. Thompson et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 321–322 (2012) 88–101

2008; Hints et al., 2010), or even extraterrestrially-driven ecosystemdisturbance (Schmitz et al., 2007) – remain under activeinvestigation.

Following the GOBE is the end-Ordovician extinction, during whichnearly 85% of this newly gained species-level diversity was lost(Jablonski, 1991; Sheehan, 2001). Although it is generally acceptedthat abrupt and widespread glaciation in the Hirnantian served as theprimary mechanism for extinction (Brenchley et al., 1994; Sheehan,2001), the timing, extent, and mechanism of global cooling remaincontroversial. Despite the temporal restriction (approximately 1 Myr)of evidence for dramatic changes in sea level and sea surface tempera-ture in the Hirnantian (Brenchley et al., 1994; Finnegan et al., 2011),evidence for glacio-eustatically driven sea level change (Pope andRead, 1998; Calner et al., 2010), expansion of cool-water depositionallithologies (Pope and Steffan, 2003; Cherns andWheeley, 2007), carbonburial and CO2 drawdown (Kump et al., 1999; Saltzman and Young,2005; Young et al., 2005, 2008, 2010; Ainsaar et al., 2010), changes inoceanic circulation (Herrmann et al., 2004; Saltzman, 2005;Thompson et al., 2010), and glacial deposition of strata in Africa(Theron, 1994; Hamoumi, 1999) suggest that climate change wasalready underway at the height of the GOBE.

Recently, hypotheses regarding extensive, low-latitude emplacementof continental basalt (Barnes, 2004) have been suggested to play a role inboth the GOBE and the climatic upheaval of the Hirnantian (Keller andLehnert, 2010; Kidder and Worsley, 2010; Lefebvre et al., 2010).Unfortunately, with the exception of a sharp drop in marine 87Sr/86Srin the Darriwilian (Qing et al., 1998; Shields et al., 2003) there is littledirect evidence for the existence of such a basaltic province. Abundantevidence for widespread bentonite deposits (Bergström et al., 1995;Kolata et al., 1996) associated with explosive volcanism in continentalmargin arc settings, however, suggest that volcanism – as well aspotential weathering of arc-related basaltic provinces –may have indeedplayed a role in moderating environmental change in the Ordovician(Young et al., 2009; Buggisch et al., 2010).

Ordovician-aged bentonites occur in North America, South America,Europe, and Asia (Huff, 2008;Huff et al., 2010), and include the Lower toMiddle Ordovician Famatinian bentonite suite of Argentina (Huff et al.,1997; Fanning et al., 2004; Astini et al., 2007), and the Upper OrdovicianDeicke–Millbrig–Kinnekulle bentonite suite of North America andSweden (Huff et al., 1992; Bergström et al., 1995, 2004; Kolata et al.,1996). In order to understand the potential role of explosive volcanismin Ordovician biodiversification and climate change, this manuscriptpresents new geochronological ages for a portion of the Famatinian K-bentonite suite and explores the timing of volcanism with respect toglobally recognized patterns of marine geochemical change.

2. Geologic background

2.1. Argentine Precordillera

The Argentine Precordillera (Fig. 1) represents a microcontinentthat rifted from the southeast margin of Laurentia in the Cambrian,drifted southward across Iapetus in the EarlyOrdovician, and eventuallydocked with Gondwana in the Late Ordovician (Thomas and Astini,1996, 1999, 2003). Biostratigraphic data for the Precordillera show aprogression from a fauna with predominantly Laurentian affinity inthe Cambrian, through a period of increased endemicity and influenceof Celtic–Baltic fauna in the Early–Middle Ordovician, to a fauna splitbetween endemic and Gondwanan affinities by the Late–MiddleOrdovician (Ramos, 1988; Benedetto, 1998, 2004; Benedetto et al.,1999; Albanesi and Ortega, 2002). Biostratigraphic data are interpretedto reflect the tectonic position of the Precordilleranmicrocontinent as itdrifted southward from Laurentia to Gondwana (Benedetto, 2004).

Sedimentary strata of the Precordillera are primarily comprised ofmixed carbonate, siliciclastic, and evaporite units deposited in a rangeof supratidal to deep subtidal environments (Astini et al., 1995). At

the base of the Precordilleran succession, evaporite-bearing shale ofthe Lower Cambrian Cerro Totora Formation marks the initial riftingof the Precordilleran terrane from southern Laurentia (Thomas andAstini, 2003), with passive margin deposition initiating with theLower to Middle Cambrian La Laja Formation (Gomez et al., 2007).Passivemargin deposition continued from theMiddle Cambrian throughthe Early Ordovician and is recorded by the dominantly shallow-marinecarbonate platformdeposition of the Zonda, La Flecha, and La Silla forma-tions (Cañas, 1999). Conformably overlying these deposits, deeper-water carbonate strata of the Lower to Middle Ordovician San Juan For-mation (Astini et al., 1995), and north–south diachronous developmentof overlying deep-water shale, mixed carbonate-shale, and carbonateof theMiddle to Upper Ordovician Gualcamayo, Las Chacritas, and Agua-ditas formations, reflect variation in subsidence rates with the approachof the Precordilleran microcontinent to Gondwana (Astini, 1995;Thomas and Astini, 1996, 2003). Finally, docking of the Precordillerawith Gondwana is marked by upper Middle to Upper Ordovician clasticwedge deposits of the Trapiche Group and its equivalents, which areinterpreted as recording development of a peripheral forebulge andsteepening of slopes to the west, in response to tectonic loading duringaccretion (Thomas and Astini, 1996, 2003).

By the Late Cambrian to Early Ordovician, an activemargin developedalong the western Gondwanan margin (Astini, 2003; Astini and Dávila,2004), and a series of Lower Ordovician K-bentonites in the Famatinaterrane show compositional evolution from volcanic arc to morecontinental affinities (Astini et al., 2007). An absence of preservedbentonites in the Precordilleran terrane at this time suggests its spatialseparation from the Gondwananmargin, with its approach to Gondwanamarked by the widespread appearance of K-bentonite deposition in theMiddle Ordovician (Huff et al., 1998; Astini et al., 2007).

Ultimately the Precordilleran terrane was uplifted by east-directedfaulting of the Andean thrust belt in the Cenozoic (Thomas and Astini,1999), and currently is exposed along the western margin of Argentina(Fig. 2), where it strikes north–south along the eastern margin of theAndean mountain range between 28°45′S and 33°15′S (Ramos, 1988,2004). To the west of the carbonate platform, deep-water turbidites,and ocean-floor mafic and ultramafic rocks were deposited (Astini et al.,

90 C.K. Thompson et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 321–322 (2012) 88–101

1995). To the east, the Famatina magmatic belt of the Sierra Pampeanasconsists primarily of igneous rocks with calc-alkaline continental marginaffinities (Astini et al., 1995). The Famatinian volcanic arc, in particular,developed on andwithin basement metasedimentary and gneissic lithol-ogies with protoliths derived, according to zircon ages, from Proterozoicto Cambrian sources (Pankhurst et al., 1998; Rapela et al., 1998).

2.2. Distribution of K-bentonites within the Precordilleran terrane

Numerous bentonite horizons have been reported in the upper SanJuan and lower Gualcamayo formations in the Argentine Precordillera(Huff et al., 1997, 1998; Fanning et al., 2004). These deposits arewidespread in the Precordillera, where they are interbedded withsubtidal, storm-dominated carbonate deposits of the upper San JuanFormation and basinal shale of the Gualcamayo/Los Azules Formation.K-bentonites have been reported from sections at Talacasto, Cerro LaChilca, Las Chacritas, Cerro Viejo, and Rio Gualcamayo sections (Huffet al., 1997, 1998, 2008; Fanning et al., 2004), and are here reported,as well, from the uppermost San Juan Formation strata near Pachaco.Bentonites are notably absent, however, from coeval, shallow-subtidaldeposits of the San Juan Formation at its type section at La Silla, likelybecause bioturbation and wave action mixed ash deposits withunderlying sediment, obscuring boundaries between carbonate deposi-tion and ash deposits. By contrast, the absence of bentonites in strata ofthe lower San Juan Formation, from all sections, likely reflects thegeographic distance of the Precordilleran terrane from the western

igneous deposits

sedimentary platform deposits

100

AN

DE

S

San Jua

Mendoza

1 Rio Gualcamayo Section

2 Cerro Viejo Section

3 Cerro La Silla Section

4 Cerro La Chilca Section

5 Talacasto Section

6 Pachaco Section

Fam

atin

a

Prec

ordi

llera

SierrPampe

Pie dePalo

Legend of Symbols

A

S. America

Arg

entin

a

Fig. 2. Geologic map of western Argentina (modified from and Keller, 1999; Thomas and Asthe Sierra Pampeanas and Famatina magmatic arcs. B) Outcrop map of Precordilleran strata1) formations. K-bentonites occur at all marked sections with the exception of the type sectsampled at Cerro La Chilca and Talacasto.

Gondwanan margin at initiation of Famatina volcanism (Astini et al.,2007).

Major and trace element analyses show K-bentonites from theTalacasto section (Fanning et al., 2004) to be high-silica rhyolites(>75% SiO2). The strikingly similar composition of bentonites collectedby Huff et al. (1998) at the Cerro Viejo section suggests lateralcontinuity of volcanic ash falls across the Precordilleran terrane.Geochemical data further demonstrate the similarity of K-bentonitesfrom both of these localities with magmas of the Chaschuil rhyolite inthe Famatinian arc system (Fanning et al., 2004), suggesting thatFamatinianmagma (468.3±3.4 Ma; Pankhurst et al., 2000) is the likelysource of these ash falls.

2.3. Stratigraphic framework of San Juan Formation bentonites

In the Argentine Precordillera, bentonites of the San Juan andGualcamayo formations were deposited in the late Floian to themid-Darriwilian (Huff et al., 1997; Fanning et al., 2004). This studyreports ages only from bentonite horizons within the San Juan Forma-tion from the Talacasto and Cerro la Chilca exposures (Fig. 2). Thesebentonites dissect a prominent, yet low magnitude, carbon-isotopeexcursion wherein the isotopic composition of marine carbonatesdrops from 0‰ to approximately −2‰ before rising again to valuesnear 0‰ (Fig. 3). This negative isotopic excursion has been correlat-ed to a near identical excursion in the lower Table Head Formation,western Newfoundland (Thompson and Kah, 2012). Overlying stra-ta of the Table Head and Table Cove formations in Newfoundland, as

20 km

2

3

Jachál

Guandacol

San Juan

Rio

San

Jua

nR

io Ja

chál

31°

30°

1

4

5

km

n

6

asanas

B

tini, 2003). A) Geographic locality of the Precordilleran sedimentary terrane relative to(colored gray), noting sections of the San Juan (localities 2–6) and Gualcamayo (localityion of the San Juan Formation at Cerro La Silla. For this study, bentonite horizons were

-3 -2 -1 0 1 2

Ear

ly O

rdov

icia

nM

iddl

e O

rdov

icia

n

Dap

ingi

anD

arri

wili

an

Seri

es

Stag

esFl

oian

Tre

mad

ocia

n

472

479

468

T.S.

*T

r1T

r2T

r3Fl

1Fl

2Fl

3D

w2

Dw

1D

p1

Dp2

Dp3

488

δ C (‰ VPDB)13

Sand

bian

Lat

e O

rd.

-3 -2 -1 0 1 2δ C (‰ VPDB)13

460.5D

w3

Sa1

Sa2 5b

5a

4c

4b

4a

3b

3a

2c

2b

2a

1d

1b

1c

1a

S.S.

*

M. dianae

P. elegans

O. evae

B. navis to O. evae

M. parva

L. variabilis

E. suecicus

P. proteus

P. deltifer

Conodont Zones

P. serra

P. anserinus

A. tvaerensis

Composite marine C-isotope curveBuggisch et al. (2003)

Composite marine C-isotope curve

Thompson and Kah (2012)Thompson et al. (2010)

La Silla Formation

San Jaun Formation

Legend of Symbols

Table Head Group, Newfoundland

San Juan Formation at Talacasto

San Juan Formation at La Chilca

Las Chacritas/Las Aguaditas Fms

San Juan and La Silla formations

La Silla Formation at La Silla

San Juan Formation at La Silla

Fig. 3. Biostratigraphic and chemostratigraphic correlation of the San Juan Formation, Argentina. A) Proposed biostratigraphic framework for carbon-isotope stratigraphy fromBuggisch et al. (2003) based on regional maximum duration of the San Juan Formation. B) Proposed biostratigraphic framework for carbon-isotope stratigraphy from Thompsonand Kah (2012), based on chemostratigraphic correlation of strata from Argentina and Newfoundland and a combination of biostratigraphic constraints from these twogeographically disparate units. Stage Slices are from Bergström et al. (2008) and Time Slices from Webby et al. (2004). Shaded box marks the positive isotope excursion knownas MDICE (Ainsaar et al., 2004; Schmitz et al., 2010). See text for discussion.


well as strata of the Las Chacritas and Las Aguaditas formations inArgentina, show continued rise in marine carbon-isotope compositionin the aftermath of this excursion to values near 1.5‰. Elevated valuesare considered equivalent to the MDICE event (Middle DarriwilianCarbon Isotope Excursion; Ainsaar et al., 2004). TheMDICE is the old-est of seven positive marine carbon-isotope excursions in the Middleto Late Ordovician (Bergström et al., 2008), and is recorded broadlyacross Baltica (Ainsaar et al., 2004, 2010; Kaljo et al., 2007), China(Schmitz et al., 2010), and potentially North America (Saltzmanand Young, 2005). North American marine carbon isotope records,however, are ambiguous (Saltzman and Young, 2005), and do notclearly show the >1‰ excursion above baseline values that definesthe MDICE excursion (Ainsaar et al., 2004; Schmitz et al., 2010).The combined dataset fromArgentina and Newfoundland, therefore, ap-pears to represent the first clear record of the MDICE excursion in thewestern hemisphere.

Bentonite deposits examined here span a low magnitude (b2‰),negative carbon-isotope excursion that is widely recognized directlyprior to the MDICE event (Ainsaar et al., 2004, 2010; Saltzman andYoung, 2005; Kaljo et al., 2007; Schmitz et al., 2010). The globalbiostratigraphic correlation of this negative isotopic excursion,however, is poorly constrained. In the Argentine Precordillera,bentonite deposits of the San Juan and Gualcamayo formations spanthe uppermost Oepikodus evae through Eoplacognathus suecicusconodont zones (Huff et al., 1998) and the victoriae to elegans grapto-lite zones (Huff et al., 1998; Astini et al., 2007), indicating an age oflate Floian (FL3) or earliest Dapingian (DP1) to late Darriwilian(DW3) (Bergström et al., 2008). Current U–Pb geochronology of thePrecordilleran K-bentonites provides ages only broadly consistentwith biostratigraphic zonation, with bentonite horizons yieldingages of 469.5±3.2 Ma and 470.1±3.3 Ma (Fanning et al., 2004) to464±2 Ma (Huff et al., 1997). Fanning et al. (2004) noted, however,


that the latter age was calculated by merging three concordant datapoints with one discordant point. A recalculated age that excludedthe discordant point, provided a mean 207Pb/206Pb age of 470.9±3.4 Ma, and suggests a late Dapingian to early Darriwilian age forthe Famatinian bentonite suite.

Across Baltica and China, the positive MDICE excursion occurslargely within the Lenodus variabilis to the Eoplacognathus suecicusconodont zones and the pre-MDICE negative isotope excursion with-in the Baltoniodus navis to Microzarkodina parva conodont zones(Ainsaar et al., 2004, 2010; Kaljo et al., 2007; Schmitz et al., 2010). Al-though regional biostratigraphic data from the Argentine Precordilleraare consistent with the pre-MDICE negative excursion occurring in theearly to mid-Darriwilian (M. parva to L. variabilis; Buggisch et al.,2003), recent correlation of carbon-isotope records from Argentinaand western Newfoundland (Thompson and Kah, 2012) shows this ex-cursion in the lower Table Head Formation (Fig. 3). Chemostratigraphicdata require that the pre-MDICE negative excursion lies substantiallybelow the top of the L. variabilis zone, which marks the top of the over-lying Table Cove Formation (Maletz et al., 2009). These correlationsplace the observed negative excursion in the Dapingian to earlyDarriwilian, which is broadly consistent with current geochronologicalconstraints, but inconsistent with the biostratigraphy of Buggisch et al.(2003). Much of the current uncertainty in the age of this chemostrati-graphic event likely arises from the regional diachroneity of the top ofthe Middle Ordovician San Juan Formation, which undergoes bothrapid deepening and differential erosion with tectonic loading and mi-gration of the peripheral bulge as the Precordilleran terrane approachesand begins to dock with the western Gondwanan margin (Thomas andAstini, 1996, 2003).

3. Geochronology of Precordilleran bentonites

3.1. Sample collection

K-bentonites were sampled from a measured section of the SanJuan Formation at Cerro La Chilca (30°36′16.9″S, 69°47′41.0″W) andTalacasto (31°00′35.5″S, 68°46′12.0″W). Carbonate samples for stableisotopic (C, S, O) analysis were collected from the same measuredsections (Thompson and Kah, 2012). The San Juan Formation at Tala-casto consists of an incomplete, composite section: lowermost strataof the San Juan Formation are deformed by chevron folds above abasal detachment and duplicated by numerous secondary faults;and the uppermost strata of the formation are erosionally removedand overlain by Silurian green shale. All samples for geochemicalanalysis were collected from the upper 125 m of the section, whichrepresent an incomplete, yet continuous and non-faulted section ofthe San Juan Formation. More than 70 bentonites appear in the mea-sured Talacasto section, and 11 were sampled for this study. Similarly,the uppermost 180 m of the San Juan Formation at the Cerro La Chilcasection were measured and sampled for stable isotopic and geochro-nological analysis. At Cerro La Chilca, bentonites are concentrated inthe upper 50 m of the formation, and in the transitional beds thatmark the boundary between the San Juan Formation and the overly-ing Gualcamayo Formation. More than 50 bentonites occur in theCerro La Chilca section, and 10 were sampled for this study. In bothlocalities, sampling was concentrated on well-preserved, relativelyundisturbed bentonite horizons with distinct, normal grading. Thesehorizons are interpreted to represent single, discrete ash fall events.Before sampling, surficial material was removed to expose the fresh-est material, and sampling focused on the lowest, most coarse-grained interval since zircons are expected to be concentrated inthese intervals.

In the sections sampled for this study (Talacasto and Cerro LaChilca; Fig. 4A, B), K-bentonites occur as discrete units that range inthickness from b1 cm to approximately 30 cm, with most rangingfrom 3 to 8 cm. K-bentonites typically are tan to buff colored, clay-

rich, and recessed relative to the host carbonate rocks. Strata underlyingbentonite horizons are iron-stained and commonly contain acephalopod-rich death assemblage. Most bentonite horizons occur assingle ash-fall deposits (Fig. 4C), although several composite K-bentonite deposits were also observed (Fig. 4D). Composite K-bentonites are identified by superposition of two or more events thatshow distinct normal grading. Grading does not occur within allhorizons, and may reflect a combination of depositional and modern(i.e. rooting, or rodent burrowing) homogenization processes.

3.2. Sample preparation and analysis

Zircon grains were dated using single grain ID-TIMS techniques(Schoene et al., 2006) at the Massachusetts Institute of Technology.Bentonites were processed using conventional magnetic and heavyliquid (methylene iodide) techniques to extract and separate zirconpopulations. Heavy mineral separates were hand picked in ethanolunder an optical microscope to separate and classify zircons basedon size, shape, clarity, and occurrence of inclusions. When possible,clear, elongate zircons with minimal inclusions were chosen forgeochronological analysis in order to analyze the most rapidlyformed, unaltered zircons. Internal zoning patterns were observedvia SEM-cathodoluminescence (Fig. 4). Grains were treated usingchemical abrasion techniques, wherein zircons were annealed inquartz beakers at ~900 °C for 60 h (Mattinson, 2005) and leached inconcentrated HF at 210 °C for 12 h to remove metamict grains ormetamict grain portions.

Once samples were chemically abraded and annealed, alladditional sample processing took place in a clean lab. Any commonPb in zircon analyses was considered to be procedural contaminationduring zircon dissolution. Zircons were transferred to Teflon beakersand cleaned by fluxing with 30% HNO3 for 30–40 min, sonication,and rinsing in ultrapure water. Initial cleaning was repeated withanother fluxing with 6.2 N HCl for 30–40 min, followed by sonicationand rinsing. Individual zircons were then dissolved in concentratedHF for 48 h in Teflon microvials at 210 °C after being spiked withEarthtime ET535 205Pb–233U–235U tracer solution.

Uranium and lead were collected by HCl-based anion exchangeprocedures modified from Krogh (1973). Pre-cleaned columns werefilled with 200–400 μm microbead resin and cleaned by flushing with6 N HCl (to flush any possible Pb contaminant), ultra-pure Milli-Qwater, 6 N HCl, 0.1 N HCl (to flush any possible U contaminant), and3 N HCl. After cleaning, dissolved zircon fractions were loaded in eachcolumn, which was then flushed with 3 N HCl three times to removezirconium. Pb was then released by flushing with 6 N HCl, and U wasreleased by flushing with 0.1 N HCl. Pb and U were collected in cleanTeflon beakers placed underneath the columns. Teflon beakers werecleaned by fluxing alternately with 6 N HCl and 6 N HF on a hot plateovernight (four times total). Pb and U were loaded together on a singleRe filament using a silica gel. U–Pb isotopic measurements wereperformed on a VG Sector-54 multi-collector thermal-ionization massspectrometer at MIT. Error and weighted means were calculated usingU–Pb Redux (McLean et al., 2008, 2009) and are reported as 2σ. Errorson the weighted mean are reported in a tripartite±X/Y/Z scheme thatdenotes, respectively, the internal laboratory error, the tracer calibra-tion error, and the tracer calibration and decay constant errors ofJaffey et al. (1971). Once ages are reported with full error calculation,they are referred to only in terms of the error that includes both thetracer calibration and decay constant errors.

4. Results and interpretation

Geochronological data from zircons of the San Juan Formation K-bentonites are summarized in Table 1. C-, S-, and O-isotope andmajor and trace element results are detailed in Thompson and Kah(2012).

Fig. 4. Outcrop occurrences of K-bentonites at Cerro La Chilca and Talacasto. A) Strata of the San Juan Formation at Cerro La Chilca, with arrows locating a number of the largerbentonite horizons. B) Uppermost strata of the San Juan Formation at Talacasto, with arrows locating a number of more or less prominent bentonite horizons. C) Two prominentK-bentonites in the upper San Juan Formation at Talacasto (see B). Bentonite horizons range from several millimeters in thickness (arrow) to approximately 20 cm in thickness.D) Composite bentonite horizon consisting of two distinct ash fall units marked by distinct, normal grading.


Heavy mineral fractions were separated from a total of nine K-bentonites (KB-1, KB-3, KB-7, KB-10, KBT-1, KBT-3, KBT-4, KBT-7,and KBT-10). These samples represent a combination of the lower-most and uppermost San Juan Formation bentonites from the LaChilca (KB samples) and Talacasto (KBT) measured sections, andtherefore maximized the potential to resolve discrete depositionalages even under conditions of high sedimentation rate. One sampleyielded no recoverable zircons (KB-10), and geochronological ana-lyses were ultimately carried out on the five samples (KB-1, KB-3,KBT-3, KBT-7 and KBT-10) that yielded the greatest abundance ofzircons. These five K-bentonite horizons also are positioned to definethe lower portion and apex of a prominent, yet lowmagnitude (~2‰)negative carbon isotope excursion that occurs prior to the MDICEcarbon isotope event (Fig. 3). KBT-3 was deposited in the lowerthird of this δ13C excursion, whereas KBT-7, KBT-10, KB-1 and KB-3were deposited as δ13C reached its negative apex and began to returnto heavier isotopic values.

4.1. Zircon morphology and textures

Zircons extracted from San Juan Formation K-bentonites aretypically small (b100 μm) and, although they show a range ofmorphologies, most are euhedral, lozenge to needle shaped, andhave bipyramidal terminations with no evidence for significanttransport. Broken grains retain sharp edges along broken surfacessuggesting recent breakage, perhaps during extraction or by handlingafter the annealing process. Zircons are generally clear, althoughsome contain distinct inclusions. Overall, grains display fine-scale,oscillatory-zoning under cathodoluminescence, with only a fewrecording more complicated internal structures such as rounded

internal cores, truncated zoning, and inclusions. Zircons with simple,zoned patterns and those that lacked cores were chosen for analyses.

Zircons from KBT-3 and KBT-7 are small (b100 μm in length) andeuhedral (Fig. 5A and B), and zircons from KBT-3 have bimodal,stubby and acicular, morphologies. We chose to analyze aciculargrains (length to width ratios >3:1) since this morphology is oftenassociated with rapid crystallization and therefore are likely to reflectan age that is closer to the eruption age (Corfu et al., 2003). Zirconsextracted from samples KBT-10 and KB-1 are generally larger(>100 μm in length), and display predominantly acicularmorphologies with length to width ratios >3:1 (Fig. 5C, D, and E).Finally, zircons from KB-3 are typically small (b100 μm in length)and euhedral, although many grains display oscillatory-zoning withinternal voids or rounded cores (Fig. 5F).

4.2. Geochronology of San Juan Formation zircons

4.2.1. Sample KBT-3NFive grains from sample KBT-3 were analyzed. Grain Z3 (Table 1)

yielded an age that is older than the other grains and, because thissample had low common Pb (0.33 pg) and the KBT-3N (needle-shapedgrain) population displayed potential for inherited cores, this age isconsidered to reflect inheritance. On a concordia plot (Fig. 6), theremaining four analyses form a group near or within uncertainty ofthe concordia curve. A weighted mean of the remaining four analysesyields a mean age of 473.45±0.40/0.49/0.70 with MSWD=0.65. Theacicular shape and oscillatory-zoning, which is typical of magmaticgrains (Corfu et al., 2003), suggest resolution at or near the eruptionage [473.45±0.70] of the volcanic ash.

Table 1U-Th-Pb data for zircons within San Juan Formation bentonites.

Samplea Composition Isotopic ratios Isotopic ages(Ma)

Thb

UPb*c

(pg)Pbcc

(pg)Pb*c

Pbc

206Pbd204Pb

206Pbe,f238U

±2σ(%)

207Pbe235U

±2σ(%)

207Pbe,f206Pb

±2σ(%)

206Pb*f,g238U

±2σabs.

207Pbg235U

±2σabs.

207Pbf,g206Pb

±2σabs.

Corr.coef.

KB-1z1 0.66 10.8 0.74 15 854 0.075446 0.098 0.5882 0.76 0.056545 0.7 468.88 0.44 469.7 2.9 474 15 0.69z2 0.71 9.77 1.46 7 396 0.075438 0.19 0.5928 1.7 0.056988 1.5 468.83 0.84 472.6 6.3 491 34 0.66z4 0.6 4.78 0.26 18 1090 0.075936 0.22 0.5955 0.75 0.056877 0.69 471.82 0.98 474.4 2.8 487 15 0.41z5 0.76 6.27 0.31 20 1149 0.075548 0.27 0.5903 1.1 0.056668 0.96 469.5 1.2 471 4 479 21 0.54z8 0.83 17 0.31 55 3031 0.075365 0.16 0.5862 0.33 0.05641 0.28 468.4 0.71 468.4 1.3 468.5 6.3 0.53z10 1.09 5.52 0.69 8 433 0.075293 0.22 0.5932 1.5 0.057139 1.4 467.96 0.99 472.9 5.8 497 32 0.43z14 1.03 8.55 0.58 15 790 0.075361 0.17 0.5908 0.97 0.056854 0.91 468.37 0.78 471.3 3.7 486 20 0.43z24 0.8 11.2 0.41 27 1536 0.075333 0.11 0.5831 0.5 0.056141 0.48 468.21 0.52 466.5 1.9 458 11 0.32z25 0.99 8.87 0.161 6 309 0.075654 0.38 0.594 2.1 0.056934 2 470.1 1.7 473.3 8 489 43 0.46

KB-3z1 0.75 45.2 0.89 51 2843 0.17883 0.062 1.8464 0.19 0.07488 0.16 1060.58 0.61 1062.1 1.2 1065.4 3.2 0.592z3 0.48 2.22 0.39 6 351 0.2294 1.8 2.674 2.4 0.0846 1.4 1331 22 1321 18 1305 26 0.821

KBT-1z1 0.71 10.2 0.26 39 2245 0.075881 0.23 0.5961 0.38 0.056972 0.55 471.5 1 474.7 1.4 490 12 −0.6z2 0.74 22.7 2.46 9 536 0.075509 0.18 0.592 1.1 0.056862 1.1 469.26 0.83 472.1 4.3 486 24 0.48z3 0.58 10.9 0.28 39 2297 0.075747 0.097 0.5909 0.43 0.056575 0.41 470.68 0.44 471.4 1.6 475 9.2 0.23z4 0.58 8.47 0.41 21 1240 0.075669 0.16 0.5886 0.78 0.05642 0.75 470.22 0.73 470 2.9 469 17 0.29z7 0.49 1.74 1.37 1 94 0.074635 1 0.532 11 0.051715 10 464 4.5 433 38 273 240 0.4z8 0.5 2.21 1.49 1 107 0.073824 0.91 0.618 6.4 0.060679 6 459.2 4 488 25 628 130 0.51z17 0.59 9.44 0.56 17 993 0.075651 0.13 0.589 0.78 0.05647 0.71 470.11 0.58 470.2 2.9 471 16 0.6z21 0.56 8.52 0.71 12 724 0.075578 0.13 0.5887 0.87 0.056495 0.82 469.68 0.57 470.1 3.3 472 18 0.48

KBT-3Nz1 0.77 3.62 0.81 4 267 0.076195 0.18 0.596 2.4 0.056714 2.2 473.37 0.82 474.6 9 480 50 0.8z2 0.83 2.35 0.38 6 360 0.076176 0.21 0.602 2.1 0.057302 1.9 473.26 0.96 478.4 7.9 503 42 0.65z3 0.82 6.36 0.33 19 1067 0.077161 1.6 0.607 1.7 0.057079 1.4 479.2 7.4 481.8 6.5 495 30 0.66z4 0.9 15.5 0.59 26 1437 0.076162 0.17 0.5949 0.53 0.056655 0.48 473.17 0.76 474 2 478 11 0.47z5 0.71 4.17 0.48 9 510 0.076274 0.16 0.593 1.4 0.056383 1.3 473.85 0.72 472.8 5.3 467 28 0.71

KBT-10z1 0.6 13.1 0.56 23 1372 0.075643 0.11 0.5904 0.47 0.056606 0.36 470.06 0.5 471.1 1.8 476.2 8 0.97z2 0.46 17 0.35 49 2980 0.075546 0.16 0.5885 0.37 0.056501 0.28 469.48 0.74 469.9 1.4 472.1 6.3 0.68z3 0.81 25.5 0.34 75 4130 0.075558 0.089 0.5894 0.31 0.05658 0.27 469.55 0.4 470.5 1.2 475.2 6 0.51z4 0.7 9.23 0.49 19 1090 0.075547 0.12 0.5902 0.7 0.05666 0.67 469.49 0.56 471 2.7 478 15 0.34z5 0.6 37.1 0.84 44 2589 0.075555 0.083 0.5887 0.26 0.056507 0.24 469.54 0.37 470.02 0.99 472.4 5.2 0.47z6 0.76 8.75 0.62 14 810 0.075953 0.12 0.6007 0.78 0.057359 0.73 471.92 0.53 477.7 3 505 16 0.49z9 0.86 8.17 0.6 14 762 0.075799 0.16 0.5946 0.91 0.056894 0.84 471 0.74 473.8 3.4 487 18 0.5z10 0.83 8.5 1.93 4 260 0.075989 0.51 0.6 2.4 0.057308 2.2 471.2 2.3 477.5 9.3 503 49 0.48

Blank composition: 206Pb/204Pb=18.24±0.21; 207Pb/204Pb=15.34±0.16; 208Pb/204Pb=37.35±0.20.a Single-grain fraction descriptors: KBT=Talacasto, KB=Cerro La Chilca, N=needles.b [Th] calculated from radiogenic 208Pb and the 207Pb and the 207Pb/235U date of the sample, assuming concordance between U–Th–Pb systems: λ (232Th)=4.9475×10−11/yr.c Pb* and Pbc represent radiogenic Pb and common Pb, respectively.d Measured ratio corrected for fractionation and spike contribution only.e Measured ratios corrected for fractionation, spike, blank and initial common Pb.f Corrected for initial Th/U disequilibrium using radiogenic 208Pb and Th/U [magma]=2.8.g Isotopic dates calculated using the decay constants of Steiger and Jager (1977): λ(235U)=9.8485×10−10/yr and λ(238U)=1.55125×10−10/yr.


4.2.2. Sample KBT-7Eight grains from sample KBT-7 were analyzed. Grains Z7 and Z8

were excluded from age calculation because of their high commonlead content (1.37 and 1.49 pg, respectively), which likely representsprocedural contamination. On a concordia plot (Fig. 6), Z1 and Z3 falllargely outside uncertainty of the concordia curve and are, therefore,excluded from the age calculation. The remaining four analyses forma group near or within uncertainty of the concordia curve. A weightedmean of these four analyses gives an age of 469.86±0.33/0.42/0.65with MSWD=1.4. Fine-scale oscillatory-zoning that is typical ofmagmatic grains (Corfu et al., 2003) suggests resolution at or nearthe eruption age [469.86±0.65] of the volcanic ash.

4.2.3. Sample KBT-10Eight grains from sample KBT-10 were analyzed. Grain Z10 was

excluded from age calculations because of its high common leadcontent (1.93 pg), which likely reflect procedural contamination.

Grains Z6 and Z9 fall well outside uncertainty of the concordiacurve (Fig. 6) and, therefore, are excluded from the age calculation.The remaining five analyses form a group near or within uncertaintyof the concordia curve. A weighted mean of these grains yields amean age of 469.63±0.21/0.33/0.60 with MSWD=0.97. The acicularshape and oscillatory-zoning, which is typical of magmatic grains(Corfu et al., 2003) suggest resolution at or near the eruption age[469.63±0.60] of the volcanic ash.

4.2.4. Sample KB-1Nine grains from sample KB-1 were analyzed. Zircons Z4, Z5 and

Z25 fall well outside uncertainty of the concordia curve (Fig. 6) and,therefore, are excluded from the age. These grains could representcontamination from inclusions. The remaining six analyses form agroup near or within uncertainty of the concordia curve. A weightedmean of these six grains yields a mean age of 469.53±0.26/0.36/0.62 with MSWD=1.2. This is interpreted as the eruption age of the

BA

DC

FE

Fig. 5. Cathodoluminescence images of San Juan Formation zircons. A) Zircon grain from sample KBT-3N (Talacasto section) showing fine-scale oscillatory zoning. B) Zircon grainfrom sample KBT-7 (Talacasto section) showing fine-scale oscillatory zoning and bipyramidal terminations. C) Fragment of acicular zircon grain from sample KBT-10 (Talacastosection) showing fine-scale oscillatory zoning. D) Acicular zircon grain from sample KBT-10 (Talacasto section) showing fine-scale oscillatory zoning and a small inclusion. E) Zircongrain from KB-1 (Cerro La Chilca section) showing oscillatory zoning that is similar to zircons from the Talacasto section. F) Zircon grain from KB-3 (Cerro La Chilca) showing sub-rounded, inherited core.


volcanic ash. The acicular shape and oscillatory-zoning, which istypical of magmatic grains (Corfu et al., 2003) suggest resolution ator near the eruption age [469.53±0.62] of the volcanic ash.

4.2.5. Sample KB-3The two grains from sample KB-3 that were analyzed each yielded

Proterozoic ages (1060 and 1330 Ma). These ages suggest that thezircons are potentially sourced from basement metasedimentaryrocks and gneisses of the Famatina and Pampeanas arc terranes(Fig. 2A), which have zircons with continental affinities that rangein age from 500 to 2000 Ma (Casquet et al., 2001; Rapela et al.,2001). Zircons analyzed from KB-3, therefore, were interpreted asinherited and no further analyses were conducted.

5. Discussion

5.1. Defining Precordilleran chronology

The resolved age of sample KBT-3 falls within the Floian Stage ofthe 2009 GSA Geologic Time Scale (Walker and Geissman, 2009);mean ages of the remaining samples fall entirely within theDapingian Stage. The ages of KBT-3, KBT-7, KBT-10, and KB-1 arewithin error of mean U–Pb zircon ages previously reported for theTalacasto section (469.5±3.2 Ma and 470.1±3.3 Ma; Fanning et al.,2004). All bentonites analyzed in this study are older than the U–Pbzircon mean age reported for the Cerro Viejo section (464±2 Ma;Huff et al., 1997), but agree well with a recalculation of this age to470.9±3.4 Ma (Fanning et al., 2004).

image of Fig.�5

KBT-3N KBT-7

KBT-10 KB-1

468

472

474

464

466

472

474

476

466

472

474

470.5

471.5

474.5

475.5

0.580 0.600 0.620 0.640

0.585 0.590 0.595 0.600 0.605

0.07

40 0

.074

5

0.07

45

0.07

50

0.07

50

0.07

55

0.07

55

0.0

760

0.07

650.

0740

0.0

745

0.0

750

0.0

755

0.0

760

0.07

65

0.0

760

0.07

65

0.580 0.585 0.590 0.595 0.600 0.605 0.610

0.580 0.585 0.590 0.595 0.600 0.605

0.07

58

0.0

760

0.

0762

0

.076

40.

0766

Pb/

U

206

238

Pb/

U

206

238

Pb/

U

206

238

Pb/

U

206

238

Z10

Z24 Z14

Z8 Z2

Z1

Z5 Z

25 Z4

6.8

25.0

11.0

13.3

9.5

34.4

Z2 Z21 Z

17 Z4 Z3

Z1

Z2

Z4

Z5

Z3 Z

1Z

9 Z6

8.2

14.3

31.8

27.6

18.1

Z4 Z2 Z1 Z5

Age (Ma)

27.6

17.4

23.6

31.3

weighted mean Pb/ U(Th-corrected)

473.45 ±0.40/0.49/0.70MSWD = 0.65, n=4

235 238


469.86 ±0.33/0.42/0.65MSWD = 1.4, n=4

235 238


469.63 ±0.21/0.33/0.60MSWD = 0.97, n=5

235 238 weighted mean Pb/ U(Th-corrected)

469.53 ±0.26/0.36/0.62MSWD = 1.2, n=6

235 238

Pb/ U207 235 Pb/ U207 235

Pb/ U207 235Pb/ U207 235

Fig. 6. U–Pb concordia diagram of whole-grain zircon analyses from the upper San Juan Formation at the Talacasto (KBT) and Cerro La Chilca (KB) sections, with internal errorwithout systematic error, tracer calibration error, and combined tracer calibration and decay constant errors. Shaded ellipses represent analyses that were included in mean agecalculations. Black bars represent weight fraction that each analyses represents in mean age calculations. Thin dotted lines represent the concordia error range resulting frompropagation of decay constant counting statistic error (2σ) from Jaffey et al. (1971).


Difficulty arises, however, when considering these ages within abiostratigraphic framework for the San Juan Formation. Thebiostratigraphic age of the San Juan Formation is well constrained atits base to the Macerodus dianae zone, which is considered coevalwith the uppermost Tremadocian Paltodus deltifer conodont zone(Albanesi et al., 1998; Buggisch et al., 2003). The top of the formation,however, is regionally diachronous, with biostratigraphic determina-tions spanning an 8–10 Myr range from conodont zones O. evae to L.variabilis (Albanesi et al., 1998, 1999; Cañas and Aguirre, 2005) or E.suecicus (Sarmiento, 1991). Diachroneity of theuppermost San JuanFor-mation is recorded through the eastern Precordillera by a north to southshift in depositional facies to deeper-water facies of the Gualcamayo, LasChacritas, and Las Aguaditas formations (cf. Keller et al., 1993; Keller,1999), or by truncation — either structurally, as at its type section of LaSilla (Buggisch et al., 2003), or stratigraphically, as at Talacasto where itis unconformably overlain by Silurian shale (Keller, 1999).

Chemostratigraphic correlation of the San Juan Formation andconformably overlying Gualcamayo, Las Chacritas, and Las Aguaditasformations with carbonate strata of the Table Point and Table Coveformations, western Newfoundland (Thompson and Kah, 2012), sug-gests a similar mid-Dapingian to mid-Darriwilian (or approximately470–464 Ma) diachroneity for the upper San Juan Formation. Insharp contrast to the correlation of biostratigraphic and chemostrati-graphic profiles (Fig. 3) proposed by Buggisch et al. (2003), the

addition of biostratigraphic markers provided by chemostratigraphiccorrelation with western Newfoundland (Thompson and Kah, 2012)requires that the uppermost strata of the San Juan Formation atboth Talacasto and Cerro La Chilca reside close to the Dapingian–Darriwilian boundary. Combined chemo-biostratigraphic correlationsof Thompson and Kah (2012) are entirely consistent with geochrono-logical data reported here. Close correspondence between the strati-graphic C-isotope profiles for Cerro La Chilca and Talacasto measuredsections suggests similar deposition rates recorded by these strata. Geo-chronological constraints support a duration of the Talacasto section(between KBT-3 and KBT-10) of 3.8±0.45 Myr, which, assuming afield measurement error of ±5%, gives a sedimentation rate of 21.8±2.8 (19 to 24.6)m/Myr. These sedimentation rates place the top of theSan Juan Formation (defined as the top of the 20.9 m thick “transitionalbeds” which mark the boundary between the San Juan and overlyingGualcamayo Formation), which lies 45.2 m above sample KB-1, at ap-proximately 467 Ma, or within the earliest Darriwilian.

5.2. Calibration of marine carbon isotopes

San Juan Formation bentonites were deposited during a promi-nent, low-magnitude (b2‰) negative carbon isotope excursion.Radiometric calibration of the San Juan Formation marine carbonisotope curve (Fig. 7) requires revision of the current biostratigraphic

-3 -2 -1 0 1

δ C (‰ VPDB)13

Fl1

Fl2

Fl3

Dp1

Dp2

Dp3

Dw

1D

w2

2a

2b

2c

3a

3b

4a

4b

Flo

ian

Dap

ingi

anD

arri

wili

an

Low

er O

rdov

icia

nM

iddl

e O

rdov

icia

n

P. elongatus-deltifer

P. elegans

O. evae

B. navis

M. parva

L. variabilis

E. suecicus

Conodont Zones

472

468

479

Seri

es

Stag

es

S.S.

*

T.S

.*

464

469.63 ± 0.60

469.86 ± 0.62

San Juan Fm - SSan Juan Fm - CTable Head Gp San Juan Fm - T

Legend of Symbols

T. laevis

469.53 ± 0.62

473.45 ± 0.70

Fig. 7. Geochronological constraints and carbon-isotope correlation of Early and Middle Ordovician strata from Argentina (San Juan Formation) and western Newfoundland (TableHead Group). A prominent, low-magnitude negative carbon-isotope event spans from the late Floian (O. evae conodont zone) to the mid-Darriwilian (uppermost L. variabilis co-nodont zone). The subsequent excursion to more positive carbon-isotope values (see Fig. 3) is interpreted to reflect represent the first clear identification of the MDICE excursion(Ainsaar et al., 2010) in both Laurentia (Western Newfoundland) and South America (Argentine Precordillera). Stage Slices are from Bergström et al. (2008) and Time Slices fromWebby et al. (2004).


interpretation (Buggisch et al., 2003). Althoughmarine carbonate rocksattributed to the San Juan Formation extend into themid-Darriwilian E.suecicus conodont zone (Sarmiento, 1991), it has been well-documented that the top of the San Juan Formation is diachronousacross the Precordillera (Albanesi et al., 1998, 1999; Cañas andAguirre, 2005). Both the Talacasto and Cerro La Chilca sections of theSan Juan Formation occur near its type locality at Cerro La Silla(Fig. 2). Chronometric calibration of the marine carbon isotope curveindicates that the top of the San Juan Formation in this region occurswithin the earliest Darriwilian. This revised calibration (Fig. 7)demonstrates that the observed negative carbon isotope event wasinitiated in the late Floian (O. evae conodont zone), reached its nadirin the mid-Dapingian (B. navis conodont zone), and returned to near0‰ values in the Darriwilian (uppermost L. variabilis to lowermost E.suecicus conodont zone)—well above the top of the San Juan Formationnear its type locality.

This revised calibration suggests, as well, the need to re-evaluatethe behavior of the marine carbon isotope record leading up to themid-Darriwilian MDICE event. Everywhere where the MDICE eventhas been established, including within Latvia and Estonia (Ainsaaret al., 2004, 2010; Kaljo et al., 2007), Sweden (Bergström, 2007;Schmitz et al., 2010), China (Schmitz et al., 2010), Newfoundland(this study), and Argentina (this study), positive carbon isotopevalues initiated near the base of the E. suecicus conodont zone andcontinue above the E. suecicus zone into the upper DarriwilianPygodus serra conodont zone. Carbon isotope values then remainelevated until the late Darriwilian (P. anserius conodont zone) beforethey begin to decline. The structure of the pre-MDICE carbon isotoperecord, however, is variable. In sections representing deposition indeep-water (deep shelf-to-slope) environments, such as the Jurmalaborehole in Latvia (Ainsaar et al., 2010) or western Newfoundland(this study; for lithologic descriptions and depositional environments,

see Thompson and Kah, 2012), marine carbon isotope records reveal abroad (6–8 Myr duration) negative excursion that reaches its nadir inthe B. navis conodont zone. By contrast, stratigraphic sections thatrepresent shallower-water (mid-shelf) environments, such as theKerguta and Mehikoorma boreholes (Kaljo et al., 2007; Ainsaar et al.,2010) and Tallinn section (Ainsaar et al., 2004) in Estonia, the Kargärdsection in Sweden (Bergström, 2007), and the Las Chacritas and LasAguaditas sections of Argentina (this study; for lithologic descriptionsand depositional environments, see Thompson and Kah, 2012), showan initial decrease in marine carbon-isotope composition followed bya distinct kick to more positive values in the L. variabilis conodontzone and an additional decline downward in the section to an isotopicnadir in the B. navis conodont zone. Sections representing depositionin environments transitional between these two states, such as the Hal-lekis Quarry section in Sweden, or the Puxi River and Maocaopu sec-tions of China (Schmitz et al., 2010), appear to record a combinationof these two signals wherein strata record a broad negative isotope ex-cursion with a small positive increase (b0.5‰) in the L. variabilis cono-dont zone.

That the pre-MDICE carbon isotope excursion appears to be quitelow in its magnitude (b2‰), long in its duration (6–8 Myr), andstrongly correlated with depositional environment suggests that theexcursion may simply reflect an isotopic depth gradient in theocean and its interaction with different marine environments. Inthis scenario the isotopic “kick” in the L. variabilis conodont zone,and increased isotopic variability represented by sections representingthe shallowest-water marine environments, such as the Tallinn sectionin Estonia (Ainsaar et al., 2004) or the Antelope Valley section in theUnited States (Saltzman and Young, 2005), may represent the inherentisotopic variability within these environments. Regardless, the marinecarbon isotope records from Argentina and Newfoundland appear toaccurately reflect open marine isotopic records as preserved


globally, suggesting that this pre-MDICE negative carbon isotope ex-cursion – despite its environmentally-controlled variation – may beused as a discrete chronologic marker.

5.3. Climate consequences of explosive volcanism in the MiddleOrdovician

One of the largest, most rapid drops in 87Sr/86Sr (~0.7088 to ~0.7078;Fig. 8) in the Phanerozoic occurs in the mid-Ordovician beginning in theupper Darriwilian P. serra conodont zone (Veizer and Compston, 1974;Burke et al., 1982; Veizer et al., 1986; Qing et al., 1998; Shields et al.,2003) and is broadly coincident with Late Ordovician explosivevolcanism, as marked by widespread bentonites deposition acrossLaurentia and Baltica (Huff et al., 1992, 1996; Bergström et al., 1995).This rapid drop has been attributed to a combination of waning Pan-African orogenic activity, decreased continental weathering associatedwith a major transgression of sea level, and an increase in either seafloor spreading (Shields et al., 2003) or weathering of young basalticprovinces associated with arc magmatism (Young et al., 2009).

Tr1

Tr2

Tr3

Fl1

Fl2

Fl3

Dp1Dp2Dp3

Dw1

Dw2

Dw3

Sa1

Sa2

Ka1

Ka2

Ka3

Ka4

Hi1Hi2

1a

1b

1c

1d

2a

2b

2c

3a

3b

4a

4b

4c

5a

5b

5c

5d

6a

6b

6c

Tre

mad

ocia

nFl

oian

Dap

ing.

Dar

riw

ilian

Sand

bian

Kat

ian

Hir.

Low

er O

rdov

icia

nM

iddl

e O

rdov

icia

nU

pper

Ord

ovic

ian

Glo

bal S

erie

s

Stag

es

Stag

e Sl

ices

*

Tim

e Sl

ices

*

-2 0 2 4 6K-bentonitedistribution

δ C (‰ VPDB)13

2

Arg KBs

Fig. 8. Timing of biospheric event in the Ordovician with respect to the distribution of explosOrdovician (Millbrig–Deicke–Kinnekulle suite). Bentonite suites have been implicated as a drelationships. Stage Slices from Bergström et al. (2008) and Time Slices from Webby et amodified from Bergström et al. (2008), sea surface temperatures (broad gray line) from Tret al. (2003), Young et al., 2009, and Thompson et al. (2010). Sea level curve from Haq and

It has further been suggested that this rapid drop in 87Sr/86Srreflected a potentially significant CO2 drop and coincident globalcooling prior to Hirnantian glaciation (Buggisch et al., 2010). Ifsignificant CO2 drawdown occurred, however, it was not recorded inthe marine sea surface temperature record. Oxygen-isotopic signaturesfrom both conodonts (Trotter et al., 2008; Herrmann et al., 2010) andclumped isotope paleothermometry (Finnegan et al., 2011) suggestthat after a protracted decline in sea surface temperature from theEarly through the Middle Ordovician (Trotter et al., 2008; Finnegan etal., 2011), sea surface temperatures stabilized in the mid-to-lateDarriwilian and remained stable until the mid-to-late Katian whentemperatures dropped dramatically with the onset of Hirnantianglaciation. Young et al. (2009) suggest that sea surface temperaturesduring this period were stabilized by explosive volcanism via a balancebetween volcanic outgassing and basalt weathering. In this scenario,late-Katian to Hirnantian cooling resulted from a combination of thecessation of volcanic outgassing associated with explosive volcanism,and continued weathering of arc-related basalts and their associatedCO2 drawdown.

Sr/ Sr

Sea Surface Temperature2 26 30 34 38 42

0.709

0

0.708

0

Sea Level

Low High

ive volcanic events in the Early–Middle Ordovician (Famatina suite, Argentina) and Lateriver to biospheric change during these intervals, yet current data do not support causall. (2004). K-bentonite distribution from Huff (2008), composite carbon isotope curveotter et al. (2008), and composite 87Sr/86Sr curve (dashed line) modified from ShieldsSchutter (2008). Gray boxes mark interval of the MDICE excursion.


In sharp contrast to theMiddle Ordovician strontium isotope record,stable, yet elevated 87Sr/86Sr values in the Early to Middle Ordovicianhave been attributed primarily to extensive weathering of Pan-Africanorogenic rocks (Qing et al., 1998; Shields et al., 2003). During the EarlyOrdovician, a small decline in marine 87Sr/86Sr (0.7090 to 0.7088;Shields et al., 2003) is broadly coincident with the onset of Famatinavolcanic activity (Fig. 8). Comparison of the timing of Argentina K-bentonite deposition to sea surface temperature records shows thatexplosive volcanism occurred near the end of a long (>25Myr) gradualdecline in sea surface temperature, with no apparent change in the rateof oceanic cooling associated with the onset of Famatinian explosivevolcanism (Huff, 2008). If a mechanism similar to that invoked byYoung et al. (2009) were operating in the Early to Middle Ordovician,gradual cooling through the Early to Middle Ordovician would requirebasaltic weathering to have consumed CO2 faster than it was producedby volcanic outgassing. Furthermore, in order to retain the markedstability and radiogenic isotopic composition of the 87Sr/86Sr record,basaltic weathering would also have to be balanced, over the longterm, by a continental weathering input. Combined, these argumentssuggest that explosive volcanism recorded by widespread bentonitesin the Argentine Precordillera played only a very limited role in globalbiospheric change.

The gradual decline in sea surface temperature observed throughthe Early and Middle Ordovician has also been attributed to anincrease in productivity and organic carbon burial as a result ofincreased biodiversification in the Ordovician (Trotter et al., 2008).Biodiversification during the GOBE is marked by transition fromdominantly benthic grazers to suspension feeders (Bottjer et al.,2001). It has been hypothesized that this transition in feedingstructure and resultant tiering of the marine ecosystem was drivenby phytoplankton expansion that served as a new food source foranimal consumption (Bambach, 1993; Signor and Vermeij, 1994). Itis thus necessary to explore whether explosive volcanism in theMiddle Ordovician, which can provide a variety of metals, such asiron and phosphorus, that are essential for organic matter production(Gaddy and Parker, 1986; Felitsyn and Kirianov, 2002), may have hadplayed a direct role in the expansion of life in the Middle Ordovician.Explosive volcanism in the Famatinian arc system appears to havebeen active through much of the early stages of the GOBE (Webbyet al., 2004). Volcanic ash deposits potentially acted as a fertilizer toenhance bioproductivity (Gaddy and Parker, 1986) during this time.If this is the case, the effects of increased nutrient input should bereflected in marine biogeochemical cycles.

Marine C-isotope composition is considered to reflect a balancebetween organic carbon burial and delivery of inorganic carbon tothe ocean. The deposition of Famatina K-bentonites coincides withan interval of unusually stable marine carbon isotope composition,wherein short-term excursions of b1‰ are superimposed over along-term variation between −2‰ and 0‰ from the mid-Floian tothe mid-Darriwilian (Fig. 8). Subdued carbon-isotope fluctuations inthe Early to Middle Ordovician are commonly attributed to generallylow rates of organic carbon burial (fraction of total carbon buried asorganic carbon=0.1 to 0.2) associated with decreased nutrientavailability during greenhouse times (Saltzman, 2005). Such aninterpretation is supported by sulfur-isotope data that indicateswidespread euxinia and the potential for a net flux of nutrients tothe sedimentary substrate (Thompson and Kah, 2012). Under theseconditions, initiation of Famatina arc volcanism in the Floian wouldenhance nutrient fluxes to the surface oceans, resulting in a net in-crease in marine productivity and organic carbon. The broad negativecarbon-isotope excursion recorded through the upper Floian andDapingian strata is inconsistent with such an interpretation. Howev-er, because marine carbon and sulfur isotope cycles are linked viathe availability of reactive marine organic matter for bacterial sulfatereduction, the sulfur isotope record may provide additional insightinto changes in marine productivity.

The isotopic composition of marine sulfate in the Early to MiddleOrdovician records short-term oscillation that has been attributed tosmall scale changes in bacterial sulfate reduction and bacterial sulfateoxidation within an ocean that otherwise maintains a dynamicequilibrium between distinct sulfate (surface ocean) and hydrogensulfide (deep ocean) reservoirs (Thompson and Kah, 2012). An abruptshift in the average isotopic composition ofmarine sulfate, however, oc-curs near the Middle–Late Ordovician boundary. This sulfur-isotopeevent coincides with a long-term minimum in sea surface temperature(Trotter et al., 2008) and has been hypothesized to represent the onsetof vigorous oceanic circulation and downwelling of cool, oxygen-richwaters that resulted in substantial oxidation of the deep-ocean hydro-gen sulfide reservoir (Thompson et al., 2010). This event and the subse-quent reorganization of marine biogeochemical cycles, however, occurwell after the end of explosive volcanism associated with Famatinianarc magmatism. Prior to this biogeochemical reorganization, the long-term stability of the marine sulfur-isotope record – which was main-tained both before and after the onset of Famatinian volcanism– suggestsno clear influence of explosive volcanism on marine environmentalconditions.

Together, these observations suggest that the relationship betweenexplosive volcanism and environmental change in the Middle Ordovi-cian is not straightforward, and that the extent of volcanism repre-sented by the Famatina bentonite suite was insufficient to affectglobal surface environments. Insteadwe suggest that expansion of phy-toplankton (Servais et al., 2008, 2010), zooplankton (Noble andDanelian, 2004; Paris et al., 2004), and ultimately marine metazoan di-versity in the earliest stages of the GOBE appears to be more closely re-lated to expansion of ecospace ultimately driven by plate tectonicreconfiguration and changes in sea level (Fig. 8). The apparent inabilityof Famatinian volcanism to affect the global biosphere may relate mostclearly to the limited spatial extent of ash falls along the western Gond-wanan margin (Huff et al., 1998; Astini et al., 2007).

6. Conclusions

Lower and Middle Ordovician strata of Argentina preserve anextensive record of explosive volcanism related to convergence of thePrecordillera terrane with the western Gondwanan margin and thedevelopment and ultimate demise of the Famatina arc system. High-resolution ID-TIMS U–Pb ages on zircons within K-bentonites of the SanJuan Formation, Argentine Precordillera provide stratigraphically consis-tent ages that range from473.45±0.70 Ma to 469.53±0.62 Ma. The sub-stantially higher precision of these new ages, and their correlation withhigh-resolutionmarine geochemical records, permit evaluation of explo-sive volcanism as an agent of global biospheric evolution.

Dated bentonites within the Famatina bentonite suite span a low-magnitude (2‰), globally-recorded negative excursion in marinecarbon isotopic composition that provides an independent mechanismfor global time-correlation. New geochronological ages are consistentwith recent carbon-isotope correlation suggesting that the San JuanFormation in the region of its type section is coeval with only the baseof the often-correlated Table Head Group of western Newfoundland.These data highlight the difficulty of using regional biostratigraphicdata – particularly within erosionally truncated or otherwise diachro-nous units – to define the time-frame of carbon-isotope chemostrati-graphy. New geochronological data also suggest that a globallyrecognized pre-MDICE negative excursion can be used as a discrete chro-nologic marker. San Juan Formation bentonites, however, cannot be dis-cretely correlated with observed, environmentally significant changesin the Middle Ordovician marine geochemical records of carbon, sulfur,strontium, or sea surface temperature, suggesting that the extent of vol-canism represented by the Famatina bentonite suite was insufficient toaffect global surface environments. These results emphasize that the rela-tionship between explosive volcanism and environmental change is notstraightforward and needs to be carefully evaluated.


Acknowledgments

Funding for this project was provided by the National GeographicSociety (NGS 7866-05 to Kah), National Science Foundation (NSF-EAR0745768 to Kah), and the American Chemical Society (ACS-PRF 48166to Kah), alongwith student grants from SigmaXi, the Geological Societyof America, and SEPM (to Thompson). We give special thanks toFernando Gomez (University of Cordoba) and Geoff Gilleaudeau(University of Tennessee) for their help in conducting field work inArgentina; and Lisa Pratt and Seth Young (Indiana University) andZhenghua Li (University of Tennessee) for their help with isotopic andelemental analyses. We also thank Michael Pope and an anonymousreviewer for comments that improved the clarity of this manuscript.

References

Achab, A., Paris, F., 2007. The Ordovician chitinozoan biodiversification and its leadingfactors. Palaeogeography, Palaeoclimatology, Palaeoecology 245, 5–19.

Ainsaar, L., Meidla, T., Tinn, O., 2004. Middle and Upper Ordovician stable isotopestratigraphy across the facies belts in the East Baltic. In: Hints, O., Ainsaar, L. (Eds.),WOGOGOB-2004 Conference Materials. Tartu University Press, Tartu, pp. 11–12.

Ainsaar, L., Kaljo, D., Martma, T., Meidla, T., Männik, P., Nõlvak, J., Tinn, O., 2010. Middle andUpper Ordovician carbon isotope chemostratigraphy in Baltoscandia: a correlationstandard and clues to environmental history. Palaeogeography, Palaeoclimatology,Palaeoecology 294, 189–201.

Albanesi, G.L., Ortega, G., 2002. Advances on conodont-graptolite biostratigraphy of theOrdovician system of Argentina. Serie Correlacion Geologica 16, 143–165.

Albanesi, G.L., Hünicken, M.A., Barnes, C.R., 1998. Bioestratigrafía de conodontes de lassecuencias ordovícicas del cerro Potrerillo, Precordillera Central de San Juan, R.Argentina. Actas XII Academia Nacional de Ciencias, Córdoba, pp. 7–72 (in Spanish).

Albanesi, G.L., Ortega, G., Barnes, C.R., Hünicken, M.A., 1999. Conodont-graptolitebiostratigraphy of the Gualcamayo Formation (Middle Ordovician) in theGualcamayo–Guandacol rivers area, Argentina Precordillera. Acta UnivesitatisCarolinae — Geologica 43, 45–48.

Astini, R.A., 1995. Geologic meaning of Arenig–Llanvirn diachronous black shales(Gualcamayo Alloformation) in the Argentine Precordillera, tectonic or eustatic?Ordovician Odyssey: Short Papers for the 7th International Symposium on theOrdovician System. Pacific Section, Society for Sedimentary Geology, Fullerton, CA,pp. 217–220.

Astini, R.A., 2003. Ordovician basins of Argentina. In: Benedetto, J.L. (Ed.), OrdovicianFossils of Argentina. Universidad Nacional de Córdoba, pp. 1–74.

Astini, R.A., Dávila, F.M., 2004. Ordovician back arc foreland and Ocloyic thrust beltdevelopment on the western Gondwana margin as a response to Precordilleraterrane accretion. Tectonics 23, TC4008. doi:10.1029/2003TC001620.

Astini, R.A., Benedetto, J.L., Vaccari, N.E., 1995. The early Paleozoic evolution of theArgentine Precordillera as a Laurentian rifted, drifted, and collided terrane: ageodynamic model. Geological Society of America Bulletin 107, 253–273.

Astini, R.A., Collo, G., Martina, F., 2007. Ordovician K-bentonites in the upper-plate activemargin of Western Gondwana, (Famatina Ranges): stratigraphic and palaeogeographicsignificance. Gondwana Research 11, 311–325.

Bambach, R.K., 1993. Seafood through time: changes in biomass, energetics, andproductivity in the marine ecosystem. Paleobiology 19, 372–397.

Barnes, C.R., 2004. Was there an Ordovician superplume event. In: Webby, B.D., Droser,M.L., Paris, F., Percival, I.G. (Eds.), The Great Ordovician Biodiversification Event.Columbia University Press, New York, pp. 77–80.

Benedetto, J.L., 1998. Early Paleozoic brachiopods and associated shelly faunas fromwesternGondwana: its bearing on the geodynamic history of the pre-Andean margin. In:Pankhurst, R.J., Rapela, C.W. (Eds.), The Proto-Andean Margin of Gondwana: GeologicalSociety of London, Special Publication, 142, pp. 57–83.

Benedetto, J.L., 2004. The allochthony of the Argentine Precordillera ten years later(1993–2003): a new paleobiologic test of the microcontinent model. GondwanaResearch 7, 1027–1039.

Benedetto, J.L., Sánchez, T.M., Carrera, M.G., Brussa, E.D., Salas, M.J., 1999. Paleontologicalconstraints on successive paleogeographic positions of Precordillera terrane duringthe early Paleozoic. Geological Society of America, Special Paper 336, 21–42.

Bergström, S.M., 2007. The Ordovician conodont biostratigraphy in the Siljan region,south-central Sweden: a brief review of an international reference standard. In:Ebbestad, J.O.R., Wickström, L.M., Högström, A.E.S. (Eds.), WOGOGOB 2007: FieldGuide and Abstracts: Sveriges Geologiska Undersökning, Rapporter ochMeddelanden,128, pp. 26–41.

Bergström, S.M., Huff, W., Kolata, D., Bauert, H., 1995. Nomenclature, stratigraphychemical fingerprinting, and real distribution of some middle Ordovician bentonitesin Baltoscandia. GFF 117, 1–13.

Bergström, S.M., Huff, W.D., Saltzman, M.R., Kolata, D.R., Leslie, S.A., 2004. The greatestvolcanic ash falls in the Phanerozoic: Millbrig and Kinnekulle K-bentonites. TheSedimentary Record 2, 4–7.

Bergström, S.M., Chen, X., Gutierrez-Marco, J.C., Dronov, A., 2008. The new chrono-stratigraphic classification of the Ordovician System and its relations to majorregional series and stages and to δ13C chemostratigraphy. Lethaia 42, 97–107.

Berner, R.A., Kothavala, Z., 2001. GEOCARB III: a revised model of atmospheric CO2 overPhanerozoic time. American Journal of Science 301, 182–204.

Bottjer, D.J., Droser, M.L., Sheehan, P.M., McGhee Jr., G.R., 2001. The ecological architectureof major events in the Phanerozoic history of marine life. In: Allmon, W.D., Bottjer,D.J. (Eds.), Evolutionary Paleoecology. The Ecological Context of MacroevolutionaryChange. Columbia University Press, New York, pp. 35–61.

Brenchley, P.J., Marshall, J.D., Carden, G.A.F., Robertson, D.B.R., Long, D.G.F., Meidla, T.,Hints, L., Anderson, T.F., 1994. Bathymetric and isotopic evidence for a short-lived Late Ordovician glaciation in a greenhouse period. Geology 22, 295–298.

Buggisch, W., Keller, M., Lehnert, O., 2003. Carbon isotope record of the Late Cambrianto Early Ordovician carbonates of the Argentine Precordillera. Palaeogeography,Palaeoclimatology, Palaeoecology 195, 357–373.

Buggisch, W., Joachimski, M.M., Lehnert, O., Bergström, S.M., Repetski, J.E., Webers, G.F.,2010. Did intense volcanism trigger the first Late Ordovician icehouse? Geology 38,327–330.

Burke, W.H., Denison, R.E., Hetherington, E.A., Koepnick, R.B., Nelson, H.F., Otto, J.B.,1982. Variation of seawater 87Sr/86Sr throughout Phanerozoic time. Geology 10,516–519.

Calner, M., Lehnert, O., Nolvak, J., 2010. Palaeokarst evidence for widespread regression andsubaerial exposure in the middle Katian (Upper Ordovician) of Baltoscandia: signifi-cance for global climate. Palaeogeography, Palaeoclimatology, Palaeoecology 296,235–247.

Cañas, F.L., 1999. Facies and sequences of the Late Cambrian–Early Ordovician carbonatesof the Argentine Precordillera: a stratigraphic comparison with Laurentian platforms.In: Ramos, V.A., Keppie, J.D. (Eds.), Laurentia–Gondwana Connections before Pangea:Geological Society of America, Special Paper, 336, pp. 43–62.

Cañas, F.L., Aguirre, H.D., 2005. High-resolution stratigraphy of the San Juan Formation,Lower Ordovician, Precordillera of Cuyo, San Juan, Argentina; preliminary results.Actas del Congresa Geologico Argentino 16, 365–370 (in Spanish).

Cardenas, A.L., Harries, P.J., 2010. Effect of nutrient availability on marine originationrates throughout the Phanerozoic eon. Nature Geoscience 3, 430–434.

Casquet, C., Baldo, E., Pankhurst, R., Rapela, C., Galindo, C., Fanning, C., Saawedra, J.,2001. Involvement of the Argentine Precordillera terrane in the Famatinian mobilebelt: U–Pb SHRIMP and metamorphic evidence from the Sierra de Pie de Palo.Geology 29, 703–706.

Cherns, L., Wheeley, J.R., 2007. A pre-Hirnantian (Late Ordovician) interval of globalcooling — the Boda event re-assessed. Palaeogeography, Palaeoclimatology,Palaeoecology 251, 449–460.

Corfu, F., Hanchar, J.M., Hoskin, P.W.O., Kinny, P., 2003. Atlas of zircon textures. In: Hanchar,J.M., Hoskin, P.W.O. (Eds.), zircon: Mineralogical Society of America, 53, pp. 468–500.

Droser, M.L., Finnegan, S., 2003. The Ordovician radiation: a follow-up to the Cambrianexplosion. Integrative and Comparative Biology 43, 178–184.

Droser, M.L., Sheehan, P.M., 1997. Palaeoecology of the Ordovician Radiation: resolutionof large-scale patterns with individual clade histories, palaeogeography andenvironments. Geobios 20, 221–229.

Fanning, C.M., Pankhurst, R.J., Rapela, C.W., Baldo, E.G., Casquet, C., Galindo, C., 2004.K-bentonites in the Argentine Precordillera contemporaneous with rhyolite vol-canism in the Famatinian Arc. Journal of the Geological Society of London 161,747–756.

Felitsyn, S.B., Kirianov, V.Y., 2002. Mobility of phosphorus during the weathering ofvolcanic ashes. Lithology and Mineral Resources 37, 275–278.

Finnegan, S., Bergmann, K., Eiler, J.M., Jones, D.S., Fike, D.A., Eisenman, I., Hughes, N.C.,Tripati, A.K., Fischer, W.W., 2011. The magnitude and duration of Late Ordovician-Early Silurian Glaciation. Science 331, 903–906.

Gaddy, A.J., Parker, R.A., 1986. Zooplankton grazing activity and assimilation in thepresence of Mount St. Helens ash. Northwest Science 60, 47–51.

Gomez, F.J., Ogle, N., Astini, R.A., Kalin, R.M., 2007. Paleoenvironmental and carbon–oxygen isotope record of Middle Cambrian carbonates (La Laja Formation) in theArgentine Precordillera. Journal of Sedimentary Research 77, 826–842.

Hamoumi, N., 1999. Upper Ordovician glaciation spreading and its sedimentary record inMoroccan North Gondwana margin. Acta Universitatis Carolinae Geologica 43, 11–14.

Haq, B.U., Schutter, S.R., 2008. A chronology of Paleozoic sea-level changes. Science322, 64–68.

Harper, D.A.T., 2006. The Ordovician biodiversification: setting and agenda for marinelife. Palaeogeography, Palaeoclimatology, Palaeoecology 232, 148–166.

Herrmann, A., Patzkowsky, M., Pollard, D., 2004. The impact of paleogeography,pCO2, poleward ocean heat transport and sea level change on global coolingduring the Late Ordovician. Palaeogeography, Palaeoclimatology, Palaeoecology206, 59–74.

Herrmann, A.D., MacLeod, K.G., Leslie, S.A., 2010. Did a volcanic mega-eruption causeglobal cooling during the Late Ordovician? Palaios 25, 831–836.

Hints, O., Delabroye, A., Nõlvak, J., Servais, T., Uutela, A., Wallin, A., 2010. Biodiversitypatterns of Ordovician marine microphytoplankton from Baltica: comparison withother fossil groups and sea-level changes. Palaeogeography, Palaeoclimatology,Palaeoecology 294, 161–173.

Huff, W.D., 2008. Ordovician K-bentonites: issues in interpreting and correlating ancienttephras. Quaternary International 178, 276–287.

Huff, W., Bergström, S.M., Kolata, D., 1992. Gigantic Ordovician ashfall in North Americaand Europe: biological, tectonomagmatic, and event stratigraphic significance.Geology 20, 875–878.

Huff, W.D., Kolata, D.R., Bergström, S.M., Zhang, Y.-S., 1996. Large-magnitude MiddleOrdovician volcanic ash falls in North America and Europe: dimensions,emplacement and post-emplacement characteristics. Journal of Volcanology andGeothermal Research 73, 285–301.

Huff, W.D., Davis, D., Bergström, S.M., Krekeler, M.P.S., Kolata, D.R., Cingolani, C., 1997.A biostratigraphicallywell-constrainedK-bentonite U–Pb zircon age of the lowermostDarriwilian State (Middle Ordovician) from the Argentine Precordillera. Episodes 20,29–33.

http://dx.doi.org/10.1029/2003TC001620


Huff, W.D., Bergström, S.M., Kolata, D.R., Cingolani, C.A., Astini, R.A., 1998. Ordovician K-bentonites in the Argentine Precordillera: relations to Gondwana margin evolution.In: Pankhurst, R.J., Rapela, C.W. (Eds.), The Proto-Andean Margin of Gondwana:Geological Society, London, Special Publication, 142, pp. 107–126.

Huff, W.D., Bergström, S.M., Kolata, D.R., 2010. Ordovician explosive volcanism. GeologicalSociety of America, Special Paper 466, 13–28.

Jablonski, D., 1991. Extinctions: a paleontological perspective. Science 253, 754–757.Jaffey, A.H., Flynn, K.F., Glendenin, L.E., Bently, W.C., Essling, A.M., 1971. Precision

measurement of the halflives and specific activities of U235 and U238. PhysicalReview C4, 1889–1906.

Kaljo, D., Martma, T., Saadre, T., 2007. Post-Hunnebergian Ordovician carbon isotopetrend in Baltoscanadia, its environmental implications and some similaritieswith thatof Nevada. Palaeogeography, Palaeoclimatology, Palaeoecology 245, 138–155.

Keller, M., 1999. Argentine Precordillera: sedimentary and plate tectonic history of aLaurentian crustal fragment in South America: Geological Society of America,Special Paper, 341 (131 pp.).

Keller, M., Lehnert, O., 2010. Ordovician paleokarst and quartz sand: evidence of volcanicallytriggered extreme climates? Palaeogeography, Palaeoclimatology, Palaeoecology 296,297–309.

Keller, M., Eberlein, S., Lehnert, O., 1993. Sedimentology of Middle Ordovician carbonatesin the Argentine Precordillera: evidence of regional relative sea-level changes.Geologische Rundschau 82, 362–377.

Kidder, D.L., Worsley, D.L., 2010. Phanerozoic Large Igneous Provinces (LIPs), HEATT(Haline Euxinic Acidic Thermal Transgression) episodes, and mass extinctions.Palaeogeography, Palaeoclimatology, Palaeoecology 295, 162–191.

Kolata, D.R., Huff, W.D., Bergström, S.M., 1996. Ordovician K-bentonites of easternNorth America. GSA Special Paper 313, 1–84.

Krogh, T.E., 1973. A low contamination method for hydrothermal decomposition ofzircon and extraction of U and Pb for isotopic age determination. Geochimica etCosmochimica Acta 37, 485–494.

Kump, L.R., Arthur, M.A., Patzkowsky, M., Gibbs, M., Pinkus, D., Sheehan, P., 1999. Aweathering hypothesis for glaciation at high atmospheric pCO2 during the LateOrdovician. Palaeogeography, Palaeoclimatology, Palaeoecology 152, 173–187.

Lefebvre, V., Servais, T., François, L., Averbuch, O., 2010. Did a Katian large igneousprovince trigger the Late Ordovician glaciation? A hypothesis tested with a carboncycle model. Palaeogeography, Palaeoclimatology, Palaeoecology 296, 310–319.

Maletz, J., Albanesi, G.L., Voldman, G.G., 2009. Lower Darriwilian radiolarians fromthe Argntine Precordillera. Geobios 42, 53–61.

Mattinson, J.M., 2005. Zircon U–Pb chemical-abrasion (CA-TIMS) method: combinedannealing and multi-step dissolution analysis for improved precision and accuracyof zircon ages. Chemical Geology 220, 47–56.

McLean, N.M., Bowring, J.F., Bowring, S.A., Schoene, R.B., 2008. More than just an age:quantitative analysis of geochronological data and uncertainty. Geological Societyof America Abstracts with Programs 40, 134.

McLean, N., Bowring, J.F., Bowring, S.A., 2009. Using statistics and software to maximizeprecision and accuracy in U–Pb geochronological measurements. AmericanGeophysical Union Abstract No. V33B-2033.

Noble, P.J., Danelian, T., 2004. Radiolarians. In: Webby, B.D., Droser, M.L., Paris, F.,Percival, I.G. (Eds.), The Great Ordovician Biodiversification Event. ColumbiaUniversity Press, New York, pp. 97–101.

Owen, A.W., Crame, J.A., 2002. Palaeobiogeography and the Ordovician and Mesozoic–Cenozoic radiations. In: Crame, J.A., Owen, A.W. (Eds.), Palaeobiogeography andBiodiversity Change: The Ordovician and Mesozoic–Cenozoic Radiations: GeologicalSociety, London, Special Publications, 194, pp. 1–11.

Pankhurst, R.J., Rapela, C.W., Casquet, C., Baldo, E., Saavedra, J., Galindo, C., 1998. EarlyPaleozoic evolution of the Gondwana margin of South America. Journal of AfricanEarth Sciences 27, 145–146.

Pankhurst, R.J., Rapela, C.W., Fanning, C.M., 2000. Age and origin of coeval TTG, I- and S-type granites in the Famatinian belt of NW Argentina. Transactions of the RoyalSociety of Edinburgh, Earth Science 91, 151–168.

Paris, F., Achab, A., Chen, X., Grahn, Y., Nolvak, J., Obut, O., Samuelsson, J., Sennikov, N.,Vecoli, M., Verniers, J., Wang, X., Winchester-Seeto, T., 2004. Chitinozoans. In:Webby, B.D., Droser, M.L., Paris, F., Percival, I.G. (Eds.), The Great OrdovicianBiodiversification Event. Columbia University Press, New York, pp. 294–311.

Pope, M.C., Read, J.F., 1998. Ordovician metre-scale cycles: implications for Ordovicianclimate and eustatic fluctuations in the central Appalachian Basin, USA. Palaeoclima-tology, Palaeogeography, Palaeoecology 138, 27–42.

Pope, M.C., Steffan, J.B., 2003. Widespread, prolonged late Middle to Late Ordovicianupwelling in North America: a proxy record of glaciation. Geology 31, 63–66.

Qing, H., Barnes, C.R., Buhl, D., Veizer, J., 1998. The strontium isotopic composition of Ordo-vician and Silurian brachiopods and conodonts: relationships to geological events andimplications for coeval seawater. Geochimica et Cosmochimica Acta 62, 1721–1733.

Ramos, V.A., 1988. The tectonics of the Central Andes : 30°–33° S latitude. In: Burchfiel,C., Suppe, J. (Eds.), Processes in Continental Lithospheric Deformation: GeologicalSociety of America, Special Paper, 218, pp. 31–54.

Ramos, V.A., 2004. Cuyania, an exotic block to Gondwana: review of a historical successand the present problems. Gondwana Research 7, 1009–1026.

Rapela, C.W., Pankhurst, R.J., Casquet, C., Baldo, E., Saavedra, J., Galindo, C., 1998. Earlyevolution of the proto-Andean margin of South America. Geology 26, 707–710.

Rapela, C.W., Pankhurst, R.J., Baldo, E., Casquet, C., Galindo, C., Fanning, C.M., Saavedra, J.,2001. Ordovician metamorphism in the Sierras Pampeanas: new U–Pb SHRIMP agesin central-east Valle Fértil and the Velasco batholith. 3° S. American Symposium onIsotope Geology, pp. 616–619.

Saltzman, M.R., 2005. Phosphorus, nitrogen, and the redox evolution of the Paleozoicoceans. Geology 33 (7), 573–576.

Saltzman, M.R., Young, S.A., 2005. Long-lived glaciation in the Late Ordovician? Isotopicand sequence-stratigraphic evidence from western Laurentia. Geology 33, 109–112.

Sarmiento, G.N., 1991. Conodonts of the Suecicus Zone (lower Llanvirnian) in the Sierra deVillicum, San Juan Precordillera, Argentina. Revista Espanola deMicropaleontologica 23,113–132 (in Spanish).

Schmitz, B., Harper, D.A.T., Peucker-Ehrenbrink, B., Stouge, S., Alwmark, C., Cronholm, A.,Bergström, S.M., Tassinari, M., Wang, X.F., 2007. Asteroid breakup linked to the GreatOrdovician Biodiversification Event. Nature Geoscience 1, 49–53.

Schmitz, B., Bergström, S.M., Xiaofeng, W., 2010. The middle Darriwilian (Ordovician)δ13C excursion (MDICE) discovered in the Yangtze Platform succession in China:implications of its first recorded occurrences outside Baltoscanadia. Journal of theGeological Society of London 167, 249–259.

Schoene, B., Crowley, J.L., Condon, D.J., Schmitz, M.D., Bowring, S.A., 2006. Reassessing theuranium decay constants for geochronology using ID-TIMS U–Pb data. Geochimica etCosmochimica Acta 70, 426–445.

Scotese, C.R., McKerrow,W.S., 1990. Revisedworldmaps and introduction. In: McKerrow,W.S., Scotese, C.R. (Eds.), Palaeozoic Palaeogeography and Biogeography: Memoir ofthe Geological Society of London, 12, pp. 1–21.

Servais, T., Lehnert, O., Li, J., Mullins, G.L., Munnecke, A., Nützel, A., Vecoli, M., 2008. The Or-dovician biodiversification: revolution in the oceanic trophic chain. Lethaia 41, 99–109.

Servais, T., Harper, D.A.T., Li, J., Munnecke, A., Owen, A.W., Sheehan, P.M., 2009.Understanding the Great Ordovician Biodiversification Event (GOBE): influences ofpaleogeography, paleoclimate, or paleoecology? GSA Today 19, 4–10.

Servais, T., Owen, A.W., Harper, D.A.T., Kröger, B., Munnecke, A., 2010. The Great OrdovicianBiodiversification Event (GOBE): the palaeoecological dimension. Palaeogeography,Palaeoclimatology, Palaeoecology 294, 99–119.

Sheehan, P.M., 2001. The Late Ordovician mass extinction. Annual Reviews of Earth andPlanetary Sciences 29, 331–364.

Shields, G.A., Carden, G.A.F., Veizer, J., Meidla, T., Rong, J.Y., Rong, Y.L., 2003. Sr, C, and Oisotope geochemistry of Ordovician brachiopods: a major isotopic event aroundthe Middle–Late Ordovician transition. Geochimica et Cosmochimica Acta 67,2005–2025.

Signor, P.W., Vermeij, G., 1994. The plankton and the benthos: origins and early historyof an evolving relationship. Paleobiology 20, 297–319.

Steiger, R.H., Jager, E., 1977. Subcomission of geochronology: convention on the use ofdecay constants in geo- and cosmochronology. Earth and Planetary Science Letters36, 359–362.

Theron, J.N., 1994. In: Williams, S.H. (Ed.), The Ordovician System in South Africa;correlation chart and explanatory notes: The Ordovician System in Greenlandand South Africa: International Union of Geological Sciences, 29, pp. 1–5.

Thomas, W.A., Astini, R.A., 1996. The Argentine Precordillera: a traveler from theOuachita embayment of North American Laurentia. Science 273, 752–757.

Thomas, W.A., Astini, R.A., 1999. Simple-shear conjugate rift margins of the ArgentinePrecordillera and the Ouachita embayment of Laurentia. Geological Society ofAmerica Bulletin 111, 1069–1079.

Thomas, W.A., Astini, R.A., 2003. Ordovician accretion of the Argentine Precordilleraterrane to Gondwana: a review. Journal of South America Earth Science 16,667–679.

Thompson, C.K., Kah, L.C., 2012. Sulfur isotope evidence for widespread euxinia and afluctuating oxycline in Early to Middle Ordovician greenhouse oceans. Palaeogeo-graphy, Palaeoclimatology, Palaeoecology 313–314, 189–214.

Thompson, C.K., Kah, L.C., Kaufman, A.J., 2010. Sulfur cycling in the late Middle Ordovician:implications for ocean circulation and the onset of Late Ordovician glaciation. GSAAbstracts with Programs 42, 513.

Trotter, J.A., Williams, I.S., Barnes, C.R., Lécuyer, C., Nicoll, R.S., 2008. Did cooling oceanstrigger Ordovician biodiversification? Evidence from conodont thermometry.Science 321, 550–554.

Veizer, J., Compston, W., 1974. 87Sr/86Sr in Precambrian carbonates as an index ofcrustal evolution. Geochimica et Cosmochimica Acta 40, 905–914.

Veizer, J., Fritz, P., Jones, B., 1986. Geochemistry of brachiopods, oxygen, and carbon iso-topic records of Paleozoic oceans. Geochimica et Cosmochimica Acta 50, 1679–1696.

Walker, J.D., Geissman, J.W., 2009. 2009 GSA geologic time scale. GSA Today 19, 60–61.Walker, L.J., Wilkinson, B.H., Ivany, L.C., 2002. Continental drift and Phanerozoic

carbonate accumulation in shallow-shelf and deep-marine settings. Journal ofGeology 110, 75–87.

Webby, B.D., Paris, F., Droser,M.L., Percival, I.G., 2004. The Great Ordovician BiodiversificationEvent. Columbia University Press, New York, pp. 1–484.

Young, S.A., Saltzman, M.R., Bergström, S.M., 2005. Upper Ordovician (Mohawkian)carbon isotope (δ13C) stratigraphy in eastern and central North America: regionalexpression of a perturbation of the global carbon cycle. Palaeogeography, Palaeo-climatology, Palaeoecology 222, 53–76.

Young, S.A., Saltzman, M.R., Bergström, S.M., Leslie, S.A., Chen, X., 2008. Paired δ13Ccarb

and δ13Corg records of Upper Ordovician (Sandbian–Katian) carbonates in NorthAmerica and China: implications for paleoceanographic change. Palaeogeography,Palaeoclimatology, Palaeoecology 270, 166–178.

Young, S.A., Saltzman, M.R., Foland, K.A., Linder, J.S., Kump, L.R., 2009. A major drop inseawater 87Sr/86Sr during the Middle Ordovician (Darriwilian): links to volcanismand climate. Geology 37, 951–954.

Young, S.A., Saltzman, M.R., Ausich, W.I., Desrochers, A., Kaljo, D., 2010. Did changes inatmospheric CO2 coincide with latest Ordovician glacial–interglacial cycles?Palaeogeography, Palaeoclimatology, Palaeoecology 296, 376–388.

Zhang, T., Shen, Y., Algeo, T.J., 2010. High-resolution carbon isotopic records from theOrdovician of South China: links to climatic cooling and the Great OrdovicianBiodiversification Event (GOBE). Palaeogeography, Palaeoclimatology, Palaeoecology289, 102–112.

palaeogeography, palaeoclimatology, palaeoecology · volcanism and regional to global environmental...

Documents